2.6 KiB
license, pipeline_tag
| license | pipeline_tag |
|---|---|
| Apache License 2.0 | text-generation |
AMchat
AM (Advanced Mathematics) Chat is a large-scale language model that integrates mathematical knowledge, advanced mathematics problems, and their solutions. This model utilizes a dataset that combines Math and advanced mathematics problems with their analyses. It is based on the InternLM2-Math-7B model and has been fine-tuned with xtuner, specifically designed to solve advanced mathematics problems.
If you find this project helpful, feel free to ⭐ Star it and help more people discover it!
Import from Transformers
To load the AMchat model using Transformers, use the following code:
from modelscope import snapshot_download, AutoTokenizer, AutoModelForCausalLM
import torch
model_dir = snapshot_download("Shanghai_AI_Laboratory/internlm2-math-7b")
tokenizer = AutoTokenizer.from_pretrained(model_dir, device_map="auto", trust_remote_code=True)
# Set `torch_dtype=torch.float16` to load model in float16, otherwise it will be loaded as float32 and might cause OOM Error.
model = AutoModelForCausalLM.from_pretrained(model_dir, device_map="auto", trust_remote_code=True, torch_dtype=torch.float16)
model = model.eval()
response, history = model.chat(tokenizer, "1+1=", history=[], meta_instruction="")
print(response)
AM (Advanced Mathematics) chat 是一个集成了数学知识和高等数学习题及其解答的大语言模型。该模型使用 Math 和高等数学习题及其解析融合的数据集，基于 InternLM2-Math-7B 模型，通过 xtuner 微调，专门设计用于解答高等数学问题。
如果你觉得这个项目对你有帮助，欢迎 ⭐ Star，让更多的人发现它！
通过 Transformers 加载
通过以下的代码加载 AMchat 模型
from modelscope import snapshot_download, AutoTokenizer, AutoModelForCausalLM
import torch
model_dir = snapshot_download("Shanghai_AI_Laboratory/internlm2-math-7b")
tokenizer = AutoTokenizer.from_pretrained(model_dir, device_map="auto", trust_remote_code=True)
# `torch_dtype=torch.float16` 可以令模型以 float16 精度加载，否则 transformers 会将模型加载为 float32，导致显存不足
model = AutoModelForCausalLM.from_pretrained(model_dir, device_map="auto", trust_remote_code=True, torch_dtype=torch.float16)
model = model.eval()
response, history = model.chat(tokenizer, "1+1=", history=[], meta_instruction="")
print(response)